Many reports have described the physiological profiles of elite athletes from different sports (2,7,9). Such investigations have consistently identified key physiological variables that are positively related to successful endurance performance, such as a high maximal oxygen uptake (˙VO2max) (5), peak power output (18), lactate threshold(14,44), and fractional utilization of˙VO2max(4). Adaptations of these variables to chronic endurance training have also been studied extensively(3,13,41).
Far less is known, however, about the effects of acute, short-term (i.e., 4- to 6-wk) intensive training interventions on athletic performance in already well-trained endurance athletes and the physiological changes that may underlie any change in physical performance. Indeed, our current ideas on appropriate training regimens are based largely on the subjective observations and experiences of coaches and athletes in the field. Sports physiologists have had limited impact on the training practices of successful competitors(48).
One form of training currently used by endurance athletes is interval, or pace/tempo, training (48). Such training typically employs sustained exercise bouts alternated with periods of slower paced activity or complete rest (10,48). Although such training is a basic element in athletic conditioning and is associated with improvements in physical performance capacity (48), little is known about the rate at which physical changes occur in response to such training.
More to the point, few studies have examined the effects of interval training on the performances of competitive athletes(1,8,11). Therefore, the primary aim of this investigation was to examine the effect of replacing a portion of aerobic“base” training (BASE) with a program of sustained, high-intensity interval training (HIT) on the performances of competitive cyclists.
Subjects. Twelve male competitive cyclists, who had at least 4 yr of endurance training and who had not performed any interval training for a minimum of 3-4 months prior to this study, acted as subjects. These men were fully informed of the risks and stresses associated with the investigation before giving their written consent to participate. The study was approved by the Research and Ethics Committee of the Faculty of Medicine of the University of Cape Town.
Anthropometry. Each subject's percent body fat was estimated from his total mass and the sum of four skin fold measurements (triceps, biceps, subscapular, and supra iliac) using the procedure described by Durnin and Womersley (12). Lean thigh volume (LTV) was estimated on the assumption that the thigh is the shape of a truncated cone(25).
During preliminary testing two subjects were excluded from the study. One became ill, and the other failed to produce consistent athletic performances during the baseline measurements. Subsequently, two more subjects failed to comply with the prescribed training program and had to be excluded from the final data analyses on the remaining (N = 8) subjects.
Baseline measurements. On at least three separate occasions during a 4- to 5-wk period prior to beginning the HIT program (described subsequently), baseline laboratory measurements were obtained on each subject. The purpose of such testing was to familiarize the subjects with the laboratory procedures and to ensure that their athletic performances were stable prior to the HIT program.
Subjects always reported to the laboratory at the same time of the day for testing and training and were instructed to “ride easily” on the day prior to such visits. The maximal incremental test and the timed ride to exhaustion (TF150) at 150% of peak sustained power output (PPO) were always performed within 72 h of the simulated 40-km time trial(TT40).
Before each test the subject's body mass (mb) was recorded to the nearest 0.1 kg. After the saddle height and handlebar position of the ergometer had been adjusted to his requirements, he began a self-paced warm-up. The duration and intensity of the warm-up was held constant for all subsequent progressive exercise tests.
Maximal incremental exercise test. All subjects first performed a maximal, sustained incremental exercise test to fatigue to determine PPO and peak O2 consumption (˙VO2peak). Cyclists performed the maximal exercise tests on an electronically braked cycle ergometer (Lode, Groningen, The Netherlands) modified with clip-in pedals and low profile racing handlebars. Power output on this ergometer was constant and independent of pedaling frequency between 60 and 120 rev·min-1. During all tests the subjects remained seated and cycled at a rate of 80-90 rev·min-1.
The progressive exercise test protocol commenced at a work rate equivalent to 3.33 W·kg-1mb. The initial exercise intensity was maintained for 150 s and then was increased by 50 W for another 150 s. After the second stage, the exercise intensity was increased by 25 W every 150 s until the subject fatigued. Fatigue coincided with either a drop in the pedaling rate of >10 rev·min-1 and/or a respiratory exchange ratio (RER) greater than 1.10 (18). The PPO was defined as the last completed work rate (W) plus the fraction of time spent in the final noncompleted work rate multiplied by 25 W(26).
During the maximal incremental exercise tests, the subjects wore a nose-clip and inspired air via a Hans Rudolph 2700 one-way valve (Vacumed, Ventura, Tucson, AZ) connected to a dry gas meter. The expired air was passed through a 15 l mixing chamber and a condensation coil to an Ametek N-22 M O2 and CD-3A CO2 analyzer (Thermox Instruments, Pittsburgh, PA). Before each test, the gas meters were calibrated with a Hans Rudolph 5530 31 syringe and the analyzers were calibrated with a 16% O2:4% CO2:80% N2 gas mixture. The instrument outputs were processed by an on-line IBM computer which calculated Vi, ˙VO2, and ˙VCO2 values using conventional equations(24).
Timed ride to fatigue (TF150). Following the maximal test, subjects rested for 15 min before undertaking a timed ride to fatigue on the same electronically braked ergometer. This test began at an exercise intensity equivalent to 2 W·kg-1mb, which was maintained for 150 s. Thereafter, the subjects were instructed to increase their cadence to ≈120 rev·min-1, and the workload was increased to 150% of their PPO. Time to fatigue was taken as the number of seconds each subject maintained a cadence of >70 rev·min-1.
40-km time trial (TT40). On a separate visit to the laboratory, all subjects performed a simulated 40-km time trial to measure athletic performance. This ride was performed under standard laboratory conditions (temperature 20°C, humidity 55%) after the subjects had been instructed to prepare for a race. During the time trial, subjects rode on their own bikes mounted on a Kingcycle ergometer (EDS Portpront Ltd., High Wycombe, Buckinghamshire, UK). The bike was attached to the ergometer by the front fork and supported by an adjustable pillar under the bottom bracket. The bottom bracket support was used to adjust the rolling resistance of the rear tire on an air-braked flywheel. Rolling resistance was set to match that of a 65-kg cyclist on a level road. Subjects performed a series of “run down” calibrations, during which they accelerated to a work rate of≈350 W and then immediately stopped pedaling while remaining seated in their normal riding position. During these calibrations the bottom bracket support was adjusted until the computer display indicated that the slowing of the flywheel matched a reference power decay curve equivalent to the rolling resistance of a 65-kg cyclist on a flat road.
From that rolling resistance and the output of a photooptic sensor monitoring the velocity of the flywheel in revolutions per second, an IBM compatible computer calculated the power output (W) that would be generated by a 65-kg cyclist riding on level terrain from the followingequation:
After the Kingcycle had been calibrated, subjects warmed up at their self-selected intensity and duration. The warm-up was kept constant for all subsequent trials. During the performance rides subjects were instructed to ride the time trial “as fast as possible” and the only feedback they received was their elapsed distance (as calculated below). During each TT40 power output, (W) and cadence were recorded every 60 s and stored by the computer for the four subsequent analyses.
Data on the reliability test of the Kingcycle ergometer were obtained from six competitive cyclists who each performed three 40-km TT. The mean (± SD) of the coefficient of variation (CV) of the times taken was 0.97 ± 0.5% (39).
From the onset of the investigation, each subject recorded his training volume and perceived training intensity in a logbook, which was subsequently used to determine his average weekly training distance and the proportion of that training to be replaced with HIT. Distances covered during the laboratory interval training sessions were estimated with the followingformula:
in which W is the power output during the HIT; h is the time spent performing HIT and ((km·h-1) W-1) is the relationship between speed and W in the simulated TT40. Interval training session distances were given to the subjects so that they could reduce their BASE training distances to keep their total training volume (km·wk-1) constant.
High-intensity interval training (HIT).Figure 1 shows the HIT protocol. HIT, which was always supervised by the same investigator, took place on six occasions during a 28-d period and consisted of 6-8 repetitions each 5 min in duration at a power output equivalent to 80% of each subject's PPO (16). Between each 5-min work bout, the subjects pedaled for 1 min at a low exercise intensity (≈100 W). After the first three interval training sessions, subjects underwent a further maximal incremental test, a TF150, and a TT40. When a subject's PPO increased as a result of the first three interval training sessions, the subsequent three training sessions were performed at 80% of the new (higher) PPO. At the completion of the HIT program, all laboratory testing was repeated.
Heart rates. Throughout all laboratory training and testing sessions, heart rates (HR) were monitored continuously by a Polar Sports Tester HR monitor (Polar Electro OY, Kempele, Finland). This monitor consisted of an electrode belt worn around the chest, a transmitter, and a wrist-mounted receiver. The receiver recorded and stored the momentary HR at 15-s intervals.
Profile of Moods State (POMS). Each week, all subjects completed the POMS inventory (32) to detect any psychological changes that may have been associated with the interval training program.
Statistical analyses. All results are presented as means ± SD. Statistical significance was assessed with a one-way analysis of variance(ANOVA) for repeated measures. When a significant F-ratio was observed, the mean differences were tested with a Duncan's New Multiple Rangepost-hoc test. A Pearson product moment correlation was used to examine relationships between variables. Results were considered significant when P was ≤ 0.05.
The physical characteristics and baseline performance data of the eight subjects who completed the investigation are shown in Table 1. There were no significant differences in mb, estimated percent body fat, LTV or HRmax following the HIT program.
Table 2 presents data for each subject's baseline PPO, TF150, and TT40 tests, along with the CV for each of the testing procedures. These values are the mean of three separate tests performed on each cyclist. The group co-efficient of variation for all baseline measurements was <1.7 ± 1.3%.
Figure 2 shows that the subjects replaced 15 ± 2% of their 305 ± 38 km·wk-1 BASE training with HIT. As intended, there was no significant difference between the total average daily training distance during BASE training and the HIT program (305 ± 38 vs 297 ± 36 km·wk-1).
Figure 3 summarizes the results of the laboratory testing procedures, from mean baseline (week 0), intermediate (week 2), and post-HIT (week 4). TF150 increased significantly after 2 wk of HIT(60.5 ± 9.3 to 67.9 ± 12.3 s; P = 0.05), but further increases in TF150 were not significantly different from the 2-wk value(Fig. 3A). PPO was not significantly greater after 2 wk(P = 0.08), but it increased significantly over the 4 wk of HIT program (416 ± 32 vs 434 ± 32 W; P = 0.01;Fig. 3B). Likewise, there was no improvement in TT40 after 2 wk of HIT, but the improvement after 4 wk was highly significant (56.4 ± 3.6 to 54.4 ± 3.2 min; P < 0.001; Fig. 3C.
These time differences resulted from an increase in the average 40-km cycling speed from 42.7 ± 2.8 to 44.2 ± 2.7 km·h-1(P < 0.001).
Figure 4 shows that post-HIT, subjects were able to sustain both a significantly higher absolute (301 ± 42 vs 326 ± 43 W; P < 0.0001; Fig. 4A) and relative(72.1 ± 6.8% vs 75.0 ± 6.8% of PPO; P < 0.05;Fig. 4B) power output for the TT40.Figure 4C and 4D show that subjects were also able to ride the TT40 at a significantly higher absolute and relative HR (166± 8 vs 170 ± 7 bpm; 89.5 ± 3.2 vs 91.6 ± 3.1% of HRpeak, both P < 0.05).
Figure 5 shows the significant (r = 0.84; P< 0.01) linear relationship between PPO (W) and 40-km cycle speed(km·h-1). The relationship between PPO and 40-km cycling speed was described by the following equation:
Improvements in PPO and TF150, however, did not correlate significantly with the increases in 40-km cycling speed.
There were no significant differences in the weekly POMS scores for any of the subjects during the entire period of investigation. POMS scores before and after HIT were 171.4 ± 41 and 160.9 ± 46.3 respectively.
The purpose of this study was to determine whether a HIT program would improve the 40-km TT performances of well-trained, competitive cyclists. To ensure that the effects of the training program could be differentiated from inherent physiological variability, it was necessary to demonstrate the reliability of the laboratory testing procedures used to evaluate changes in physical performance. We (39) have recently shown that laboratory simulated 40-km time trials are highly reproducible (CV 1.0± 0.5%) in well-trained cyclists. Others (19) have also shown that trained cyclists are consistently able to reproduce TT40 performances in a laboratory setting when the influences of external factors unrelated to the exercise test are properly controlled. Before any experimental intervention took place in this study, each laboratory test was undertaken a minimum of three times. The results of these tests revealed that the mean CV (expressed as the mean of each subject's individual CV) was < 1.7% for both the PPO and TF150, and < 1.0% for the TT40(Table 2). Hickey et al.(19) recommend that the minimum acceptable changes due to experimental intervention should exceed 2.4 and 1.0% for work bouts of≈1 min and 1 h, respectively. In this study the percentage improvements in performance from week 0 to week 4 all exceeded such criteria.
Hence, the first relevant finding of this study was that when highly trained cyclists (VO2peak 5.2 I·min-1) replaced a portion(15 ± 2%) of their prolonged, moderate-intensity BASE training with sustained, intense aerobic interval training (Fig. 2), there was a significant improvement in 40-km TT performance(Fig. 3C). This improvement was a result of an increase in both the absolute and relative power output subjects could sustain after the HIT program (Fig. 4A and B). The improvement in work rate (from 301 ± 42 to 326 ± 43 W) enabled cyclists to ride the TT40 at an average speed of 44.2 ± 2.7 km·h-1 compared to 42.7 ± 2.8 km·h-1 before the interval training program. Coyle et al. (9) have previously reported that “elite national class cyclists” with similar(≈5.0 I·min-1) ˙VO2max values and 40-km TT performances (51-56 min) to those of our subjects can maintain a similarly high average work rate (≈350 W) for a 60-min “performance ride.” The average work rate of their subjects corresponded to ≈90% of ˙VO2max, whereas for our subjects it corresponded to ≈85% of˙VO2max.
The second finding was that the intensive training program also resulted in significant increases in PPO (Fig. 3B). We(18) and others (26,45) have reported highly significant relationships between the peak workload an athlete can attain during a maximal incremental test to exhaustion and˙VO2max. Therefore, the large (≈5%) increase in PPO measured in our subjects is consistent with a corresponding increase in˙VO2max. Although many investigations have reported an increase in˙VO2max following a program of endurance training(15,20,21), such studies, for the most part, have used previously sedentary individuals(29,30) or moderately trained recreational athletes (15,20,27). To our knowledge only one previous investigation (1) has examined the physiological adaptations that result from an increase in training intensity in previously well conditioned athletes. Acevedo and Goldfarb(1) found that 8 wk of high-intensity (85-90% of˙VO2max) interval training (3 d·wk-1) improved both 10-km race performance and exercise time to fatigue at a supramaximal running speed in seven competitive long-distance runners. Their intensified training program did not result in any improvements in the runners' ˙VO2max, indicating that improvements in athletic performance can be independent of increases in ˙VO2max in both moderately (11) and highly trained athletes (6,8,31).
Coyle et al. (9) observed that 40-km TT performance was closely related to the average absolute power output their cyclists were able to maintain during a maximal 60-min laboratory performance test, which is not surprising in view of the similar duration of their laboratory and criterion performance tests. However, this study shows that a significant relationship exists between the PPO the cyclists could attain during the maximal incremental test and the speed they could sustain during the TT40(Fig. 5), indicating that a laboratory test of relatively short duration (≈10 min) can predict exercise performance of much longer duration (≈1 h). We (38,43) and others (36,42) have shown that peak treadmill velocity measured during a maximal test predicts running performances at distances from 5 to 90 km. Yet, somewhat surprisingly, we were unable to demonstrate a significant relationship between either the increase in a cyclist's PPO or his maximal short-term (≈60 s) exercise capacity and his corresponding improvement in TT40.
The third finding of this study was that HIT produced a significant improvement in muscular resistance to fatigue as measured by TF150(Fig. 3A). This improvement in “anaerobic” performance became statistically significant earlier than the improvements in PPO and TT40. Indeed, PPO was not significantly greater after 2 wk but increased significantly over the 4 wk of HIT program (Fig. 3B).
Previous studies using high-intensity, short-duration (5-30 s) multiple sprint training programs have reported increases in maximal (5 s) power(28,37,46) and short-term (<30 s) performance (22,34). However, to our knowledge this is the first investigation to find improvements in maximal exercise lasting ≈60 s after a program of predominantly aerobic interval training. Although these results may seem surprising since the duration of the TT150 test differs markedly from the duration of the sustained (5 min) work bouts employed during the HIT, recent data(33,47) indicate that the contribution from aerobic energy release to the total energy requirements of 60 s of maximal work is approximately 50%. Thus, the HIT program used in this investigation would be expected to result in improvements in exercise time to fatigue when the contribution from aerobic energy is a significant component of the exercise task (for review see (17).
The demands of a sudden increase in training load over several days or weeks have failed to improve (8) or have led to a decrease in performance (23), which is sometimes accompanied by a disturbance in mood state(35,40). However, this study showed there was no significant disturbances in the athletes' emotional adaptation to the intense interval training program. The mood states of the athletes remained stable throughout the trial.
An issue that needs to be acknowledged is the impact that any type of experimental intervention may have upon human exercise performance. Athletes are very suggestive to novel training techniques, especially interval training, which they already believe is “supposed to” improve performance. The extent to which physiological versus psychological factors may be responsible for the improvement in performance is not easily determined.
In conclusion, the results of this study show that 4 wk of high-intensity interval training significantly improved the athletic performances of highly trained cyclists in laboratory tests ranging from ≈60 s to ≈1 h. The precise mechanisms underlying these improvements remain to be determined.
1. Acevedo, E. O. and A. H. Goldfarb. Increased training intensity effects on plasma lactate, ventilatory threshold, and endurance.Med. Sci. Sports Exerc.
2. Burke, E. R., F. Cerny, D. Costill, and W. Fink. Characteristics of skeletal muscle in competitive cyclists. Med. Sci. Sports Exerc.
3. Clausen, J. P. Effect of physical training on cardiovascular adjustments to exercise
in man. Physiol. Rev.
4. Costill, D. L. Physiology of marathon running.J.A.M.A.
5. Costill, D. L. and E. Winrow. Maximal oxygen intake among marathon runners. Arch Phys. Med. Rehabil.
6. Costill, D. L. The relationship between selected physiological variables and distance running performance. J. Sports Med. Phys. Fitness
7. Costill, D. L., W. J. Fink, and M. L. Pollock. Muscle fiber composition and enzyme activities of elite distance runners. Med. Sci. Sports Exerc.
8. Costill, D. L., M. G. Flynn, J. P. Kirwan, et al. Effects of repeated days of intensified training on muscle glycogen and swimming performance. Med. Sci. Sports Exerc.
9. Coyle, E. F., M. E. Feltner, S. Kautz, et al. Physiological and biochemical factors associated with elite endurance cycling performance. Med. Sci. Sports Exerc.
10. Daniels, J. and N. Scardina. Interval training and performance. Sports Med.
11. Daniels, J. T., R. A. Yarbrough, and C. Foster. Changes in ˙VO2max
and running performance with training. Eur. J. Appl. Physiol.
12. Durnin, J. V. G. A. and J. Womersley. Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br. J. Nutr.
32: 77-97, 1974.
13. Ekblom, B. Effect of physical training on the oxygen transport system in man. Acta Physiol. Scand.
14. Farrell, P. A., J. H. Wilmore, J.H., E. F. Coyle, J. E. Billing, and D. L. Costill. Plasma lactate accumulation and distance running performance. Med. Sci. Sports Exerc.
15. Gaesser, G. A. and L. A. Wilson. Effects of continuous and interval training on the parameters of the power-endurance time relationship for high intensity exercise
. Int. J. Sports Med.
16. Hawley, J.A. State of the art training guidelines for endurance performance. New Zealand Coach
17. Hawley, J. A. and W. G. Hopkins. Aerobic glycolytic and aerobic lipolytic power systems: a new paradigm with implications for endurance and ultra endurance events. Sports Med.
18. Hawley, J.A. and T. D. Noakes. Peak power output predicts maximal oxygen uptake and performance. Eur. J. Appl. Physiol.
19. Hickey, M. S., D. L. Costill, G. K. McConnell, J. J. Widrick, and H. Tanaka. Day-to-day variation in time trial
cycling performances. Int. J. Sports Med.
20. Hickson, R. C., J. M. Hagberg, A. A. Eshani, and J. O. Holloszy. Time course of the adaptive responses of aerobic power and heart rate to training. Med. Sci. Sports Exerc.
21. Hurley, B. F., J. O. Hagberg, and W. K. Allen. Effects of training on blood lactate levels during submaximal exercise
. J. Appl. Physiol.
22. Jenkins, D. G., S. Brooks, and C. Williams. Improvements in multiple sprint ability with three weeks of training.Int. J. Sports Med.
23. Jeukendrup, A. E., M. K. C. Hesselink, A. C. Snyder, and H. A. Kuipers. Physiological changes in male competitive cyclists after two weeks of intensified training. Int. J. Sports Med.
24. Jones, N.L. and E. J. M. Campbell. Calculation of results (Appendix A). In: Clinical Exercise Testing
, London: W. B. Saunders, pp. 235-239, 1982.
25. Katch, V. L. and F. I. Katch. A simple anthropometric method for calculating segmental leg limb volume. Res. Q.
45: 211-214, 1974.
26. Kuipers, H., F. T. J. Verstappen, H. A. Keizer, P. Guerten, and G. van Kraneberg. Variability of aerobic performance in the laboratory and its physiological correlates. Int. J. Sports Med.
27. Knuttgen, H. G., L. O. Nordesjo, B. Ollander, and B. Saltin. Physical conditioning through interval training with young male adults. Med. Sci. Sports Exerc.
28. Linossier, M. T., C. Denis, D. Doromis, A. Geyssant, and J. R. Lacour. Ergometric and metabolic adaptation to a 5-s sprint training programme. Eur. J. Physiol.
29. Lortie, G., J. A. Simoneau, P. Hamel, M. R. Boulay, F. Landry, F. and C. Bouchard. Responses of maximal aerobic power and capacity to aerobic training. Int. J. Sports Med.
30. Mackrides, L., G. J. F. Heigenhauser, and N. L. Jones. High intensity endurance training in 20- to 30- and 60- to 70-yr-old healthy men. J. Appl. Physiol.
31. Martin, D. E., D. H. Vroon, D. F. May, and S. P. Pilbeam. Physiological changes in elite male distance runners training.Physician Sportsmed.
32. McNair, D. N. and M. Lorr. Droppleman: Profile of Mood States Manual
. San Diego, CA: Educational and Industrial Testing Service, 1971.
33. Medbo, J. I. and I. Tabata. Relative importance of aerobic and anaerobic energy release during short-lasting exhausting bicycle exercise
. J. Appl. Physiol.
34. Medbo, J. I. and S. Burgers. Effects of training on the anaerobic capacity. Med. Sci. Sports Exerc.
35. Morgan, W. P., D. R. Brown, J. S. Raglin, P. J. O'Connor, and K. A. Kllickson. Psychological monitoring of over training and staleness. Br. J. Sports Med.
36. Morgan, D. W., F. D. Baldini, P. E. Martin, and W. M. Kohrt. Ten kilometer performance and predicted velocity at ˙VO2max
among well-trained male runners. Med. Sci. Sports Exerc.
37. Nevill, M. E., L. H. Boobis, S. Brooks, and C. Williams. Effect of training on muscle metabolism during treadmill sprinting.J. Appl. Physiol.
38. Noakes, T. D., K. H. Myburgh, and R. Schall. Peak treadmill running velocity during the ˙VO2max
test predicts running performance. J. Sports Sci.
39. Palmer, G. S., S. C. Dennis, T. D. Noakes, and J. A. Hawley. Assessment of the reproducibility of performance testing on an air-braked cycle ergometer. Int. J. Sports Med.
40. Raglin, J. S., W. P. Morgan, and P. J. O'Connor. Changes in mood states during training in female and male college swimmers.Int. J. Sports Psychol.
41. Saltin, B. Physiological effects of physical conditioning. Med. Sci. Sports Exerc.
42. Scott, B. K. and J. A. Houmard. Peak running velocity is highly related to distance running performance. Int. J. Sports Med.
43. Scrimgeour, A. G., T. D. Noakes, B. Adams, and K. Myburgh. The influence of weekly training distance on fractional utilization of maximum aerobic capacity in marathon and ultra marathon runners.Eur. J. Appl. Physiol.
44. Sjodon, B. and I. Jacobs. Onset of blood lactate accumulation and marathon running performance. Int. J. Sports Med.
45. Storer, T. W., J. A. Davis, and V. J. Caiozzo. Accurate prediction of ˙VO2max
in cycle ergometry. Med. Sci. Sports Exerc.
46. Thorstensson, A., B. Sjodin, and J. Karlsson. Enzyme activities and muscle strength after “sprint training” in man.Acta Physiol. Scand.
47. Withers, R. T., W. M. Sherman, D. G. Clark, et al. Muscle metabolism during 30, 60, and 90 s of maximal cycling on an air-braked ergometer. Eur. J. Appl. Physiol.
48. Wells, C. L. and R. R. Pate. Training for performance of prolonged exercise
. In: Perspectives in Exercise Science and Sports Medicine
Vol. 1, D. R. Lamb and R. Murray (Eds.). Indianapolis: Benchmark Press, 1988, pp. 357-391.